标题： 在保持相似性的稀疏随机嵌入中的设计选择 显示英文标题

标题： On Design Choices in Similarity-Preserving Sparse Randomized Embeddings

摘要： 扩展与稀疏化是一个在解剖结构相似的神经回路中观察到的原则，这些回路存在于蘑菇体（昆虫）和小脑（哺乳动物）中。感官数据被随机投射到更高维度（扩展部分），其中只有少数最强烈兴奋的神经元被激活（稀疏化部分）。这一原则已被用于设计FlyHash算法，该算法形成保留相似性的稀疏嵌入，已被发现对新颖性检测、模式识别和相似性搜索等任务有用。尽管其简单性，FlyHash有许多需要设置的设计选择，例如输入数据的预处理、稀疏化激活函数的选择以及随机投影矩阵的形成。在本文中，我们探讨这些选择对使用FlyHash嵌入进行相似性搜索性能的影响。我们发现，设计选择的正确组合可能导致搜索性能的显著差异。 显示英文摘要

摘要： Expand & Sparsify is a principle that is observed in anatomically similar neural circuits found in the mushroom body (insects) and the cerebellum (mammals). Sensory data are projected randomly to much higher-dimensionality (expand part) where only few the most strongly excited neurons are activated (sparsify part). This principle has been leveraged to design a FlyHash algorithm that forms similarity-preserving sparse embeddings, which have been found useful for such tasks as novelty detection, pattern recognition, and similarity search. Despite its simplicity, FlyHash has a number of design choices to be set such as preprocessing of the input data, choice of sparsifying activation function, and formation of the random projection matrix. In this paper, we explore the effect of these choices on the performance of similarity search with FlyHash embeddings. We find that the right combination of design choices can lead to drastic difference in the search performance.